Neural bases of Frustration-Aggression Theory: A multi-domain meta-analysis of functional neuroimaging studies

Abstract

Early evidence suggests that unexpected non-reward may increase the risk for aggressive behaviors. Despite the growing interest in understanding brain functions that may be implicated in aggressive behaviors, the neural processes underlying such frustrative events remain largely unknown. Furthermore, meta-analytic results have produced discrepant results, potentially due to substantial differences in the definition of anger/aggression constructs. Therefore, coordinate-based meta-analyses on unexpected non-reward and retaliatory behaviors in healthy subjects were conducted. Conjunction analyses were further examined to discover overlapping brain activations across these meta-analytical maps. Frustrative non-reward deactivated the orbitofrontal cortex, ventral striatum and posterior cingulate cortex, whereas increased activations were observed in midcingulo-insular regions, as well as dorsomedial prefrontal cortex, amygdala, thalamus and periaqueductal gray, when using liberal threshold. Retaliation activated of midcingulo-insular regions, the dorsal caudate and the primary somatosensory cortex. Conjunction analyses revealed that both strongly activated midcingulo-insular regions. Our results underscore the role of anterior midcingulate/pre-supplementary motor area and fronto-insular cortex in both frustration and retaliatory behaviors. A neurobiological framework for understanding frustration-based impulsive aggression is provided.

Systematic Review and Meta-analysis: Task-based fMRI Studies in Youths With Irritability

“…Therefore, such spatial convergence might not generalize to developmental clinical groups (childhood irritability in this case) in which neural convergence is a complex function of developmental changes, symptom variations, and compensatory neural mechanisms. 32,69 Regarding other relevant phenotypes, a recent meta-analysis in youths with depression or anxiety disorders (k=48 70 ) reported increased activation in the bilateral amygdala (especially for anxiety disorders) across a range of emotion regulation and decision-making tasks, which extended to the anterior cingulate and putamen⎯regions in the salience network that were also reported in some of the individual irritability studies reviewed here. However, no significant clusters were found when restricting to youths with major depressive disorder (MDD), suggesting that the lack of neural convergence may not be an issue specific to the irritability phenotype.…”

“…Therefore, such spatial convergence might not generalize to developmental clinical groups (childhood irritability in this case) in which neural convergence is a complex function of developmental changes, symptom variations, and compensatory neural mechanisms. 49 , 69 …”

“…For instance, two meta-analyses found that state anger (k=39 67 ) and anger experience (k=26 68 ) were associated with activations in the anterior insula/inferior frontal gyrus and anterior cingulate cortex, 67,68 and in the ventrolateral prefrontal cortex. 68 In a more relevant meta-analysis (k=68 69 ), frustrative nonreward processing was associated with deactivation in the orbitofrontal cortex, posterior cingulate cortex, and ventral striatum, and heightened activation in the midcingulo-insular regions. It is thought that the deactivation patterns represent frontal neural deficiency, which disinhibits aggressive responding to frustrative events.…”

“…It is thought that the deactivation patterns represent frontal neural deficiency, which disinhibits aggressive responding to frustrative events. 69 These meta-analyses, however, were performed in primarily healthy young adults, with a larger collection of studies employing more homogeneous tasks. Therefore, such spatial convergence might not generalize to developmental clinical groups (childhood irritability in this case) in which neural convergence is a complex function of developmental changes, symptom variations, and compensatory neural mechanisms.…”

Towards a Neurobiologically-driven Ontology of Mental Functions: A Data-driven Summary of the Twenty Years of Neuroimaging Meta-Analyses

“…However, the deactivation pattern is mainly task-dependent (Spreng, 2012). For example, when one does not receive an expected reward (frustrative non-reward), the nucleus accumbens and vmPFC seems to deactivate (Dugré & Potvin, 2021), which mirrors a diminished release in dopamine during reward omission (Schultz, 2016;Sosa, Mata-Luévanos, & Buenrostro-Jáuregui, 2021;Tian & Uchida, 2015). Further investigation is needed to determine whether two similar brain maps (based on activation), such as the multiple-demand and cognitive control networks, could be better distinguished by their pattern of deactivation.…”